International aid may take the form of multilateral aid – provided through international bodies such as the UN, or NGOs such as Oxfam – or bilateral aid, which operates on a government-to-government basis. There is considerable debate about whether international aid works, in the sense of reducing poverty and stimulating development.

However, the effectiveness of aid is often diluted by corruption. Aid is invariably channeled through the governments of recipient countries, in which power is often concentrated in the hands of a few politicians and bureaucrats, and the mechanisms of accountability are, at best, poorly developed. This tends to benefit corrupt leaders and elites rather than the people, projects and programs for which it was intended.

The hypothesis that foreign aid can promote growth in developing countries was explored, using panel data series for foreign aid, while accounting for regional differences in Asian, African, Latin American, and the Caribbean countries as well as the differences in income levels, the results of this study also indicate that foreign aid has mixed effects on economic growth in developing countries.

This study examines the relationships between foreign aid, institutional structure, and economic performance for 80 countries in Europe, America, Africa, and Asia. It is found that official development assistance and the quality of institutional structure in the sample countries affect economic growth positively.

Cargando Librerias

Algunas librerias y paquetes usados para obtener y descargar los datos

library(tidyverse) # manejo de dataframes
## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──
## ✓ ggplot2 3.3.5     ✓ purrr   0.3.4
## ✓ tibble  3.1.2     ✓ dplyr   1.0.7
## ✓ tidyr   1.1.3     ✓ stringr 1.4.0
## ✓ readr   2.0.1     ✓ forcats 0.5.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag()    masks stats::lag()
library(WDI)       # libreria para acceder a metadata de banco mundial
library(readxl)    # leer archivos de excel
library(readr)     # leer archivos csv
library(visdat)    # visualizacion de datos como graficos
library(plotly)    # graficos
## 
## Attaching package: 'plotly'
## The following object is masked from 'package:ggplot2':
## 
##     last_plot
## The following object is masked from 'package:stats':
## 
##     filter
## The following object is masked from 'package:graphics':
## 
##     layout
library(purrr)     # funcion map
library(plm)       # modelos lineales para datos panel
## 
## Attaching package: 'plm'
## The following objects are masked from 'package:dplyr':
## 
##     between, lag, lead

Obtener datos

Datos para paises bajos ingresos sean utilizados, segun clasificación del banco mundial, hay 26 paises de bajos ingresos y 51 de ingresos medios bajos

country_class <- read_excel("CLASS.xlsx")

country_class %>%
  filter(!is.na(Region), !is.na(`Income group`)) %>%
  group_by(`Income group`) %>%
  summarise(countries = n())

Listado de paises a analisar:

my_countries <- country_class %>%
  filter(!is.na(Region), `Income group` %in% c('Low income', 'Lower middle income')) %>%
  select(Code)
my_countries

Hacer la respectiva asociacion de nombres iso3c e iso2c

my_countries$iso2c <- WDI_data$country %>%
  filter(iso3c %in% my_countries$Code) %>%
  .$iso2c
my_countries

Datos del banco mundial (para ODA y los indices de gobernanza) y el Human Development Reports API son descargados desde scripts de Python. Son almacenados en archivos CSV y luego son cargados aqui:

HDI

datos_HDI <- read_csv("datos_python_HDI.csv", col_names = c('Code', 'iso2c', 'indicator', 'year', 'value'), 
                      col_types = list(col_character(), col_character(), col_character(), col_double(), col_double()))

hdi_indicators <- datos_HDI %>% distinct(indicator) %>% .$indicator

ODA

oda_indicators <- c(
'DT_ODA_ALLD_CD',
'DT_ODA_ALLD_KD',
'DT_ODA_OATL_CD',
'DT_ODA_OATL_KD',
'DT_ODA_ODAT_CD',
'DT_ODA_ODAT_GI_ZS',
'DT_ODA_ODAT_GN_ZS',
'DT_ODA_ODAT_KD',
'DT_ODA_ODAT_MP_ZS',
'DT_ODA_ODAT_PC_ZS',
'DT_ODA_ODAT_XP_ZS'
)
gob_indicators <- c(
'CC_EST',
'CC_NO_SRC',
'CC_PER_RNK',
'CC_PER_RNK_LOWER',
'CC_PER_RNK_UPPER',
'CC_STD_ERR',
'GE_EST',
'GE_NO_SRC',
'GE_PER_RNK',
'GE_PER_RNK_LOWER',
'GE_PER_RNK_UPPER',
'GE_STD_ERR',
'PV_EST',
'PV_NO_SRC',
'PV_PER_RNK',
'PV_PER_RNK_LOWER',
'PV_PER_RNK_UPPER',
'PV_STD_ERR',
'RQ_EST',
'RQ_NO_SRC',
'RQ_PER_RNK',
'RQ_PER_RNK_LOWER',
'RQ_PER_RNK_UPPER',
'RQ_STD_ERR',
'RL_EST',
'RL_NO_SRC',
'RL_PER_RNK',
'RL_PER_RNK_LOWER',
'RL_PER_RNK_UPPER',
'RL_STD_ERR',
'VA_EST',
'VA_NO_SRC',
'VA_PER_RNK',
'VA_PER_RNK_LOWER',
'VA_PER_RNK_UPPER',
'VA_STD_ERR'
)
gdp_indicators <- c(
'NY_ADJ_NNTY_PC_CD',
'NY_ADJ_NNTY_PC_KD',
'NY_ADJ_NNTY_PC_KD_ZG',
'NY_GDP_PCAP_CN',
'NY_GDP_PCAP_KN',
'NY_GDP_PCAP_CD',
'NY_GDP_PCAP_KD',
'NY_GDP_MKTP_KD_ZG',
'NY_GDP_DEFL_ZS_AD',
'NY_GDP_DEFL_ZS',
'NY_GDP_MKTP_CD',
'NY_GDP_MKTP_CN',
'NY_GDP_MKTP_KN',
'NY_GDP_MKTP_KD',
'NY_GDP_PCAP_KD_ZG',
'NY_GDP_PCAP_PP_KD',
'NY_GDP_PCAP_PP_CD',
'SL_GDP_PCAP_EM_KD',
'SP_POP_GROW'
)

datos_WB <- data.frame(indicator = character(), iso2c = character(), year = double(), value = double())

suppressWarnings(
  for (indicator in c(oda_indicators, gob_indicators, gdp_indicators)) {
    datos_WB <- rbind(datos_WB, read_csv(paste("datos_python", indicator, ".csv", sep =''), 
                                           col_names = c('indicator', 'iso2c', 'year', 'value'),
                                           col_types = list(col_character(), col_character(), col_double(), col_double())))
  }
)

Manipulacion de Datos

Transformar la estructura de los datos para una mejor comprension

datos_paper <- rbind(datos_WB, datos_HDI %>% select(indicator, iso2c, year, value)) %>%
  pivot_wider(names_from = indicator, values_from = value)

Revisar que datos estan como faltantes

vis_dat(datos_paper %>% select(all_of(gsub("_", ".", oda_indicators)))) 

  # DT.ODA.OATL.CD and DT.ODA.OATL.KD faltan
  # DT.ODA.ODAT.GI.ZS, DT.ODA.ODAT.GN.ZS, DT.ODA.ODAT.MP.ZS and DT.ODA.ODAT.XP.ZS tienen faltas
  # Un par de ocurrencias pais-año que faltan datos

vis_dat(datos_paper %>% select(NY.GDP.PCAP.CN, NY.GDP.PCAP.CD)) 

  # NY.GDP.PCAP.CN, NY.GDP.PCAP.CD, NY.GDP.MKTP.CD, NY.GDP.MKTP.CN son buenos candidatos, 
  # 'SY'falta PIB per Capita en 2022, 2023 sin datos algunos paises

vis_dat(datos_paper %>% arrange(year) %>% select(all_of(gsub("_", ".", gob_indicators)))) 

  # Datos del 2000 para atras tienen espacios faltantes 

vis_dat(datos_paper %>% select(all_of(hdi_indicators))) 

  # abr, co2_prod, le, le_f, le_m, mmr son las pocas categorias sin datos faltantes

vis_dat(datos_paper %>% arrange(year) %>% select(hdi)) 

  # hdi faltante en multiples ocaciones

vis_dat(datos_paper %>% arrange(iso2c) %>% select(SP.POP.GROW)) 

  # ZW no tiene datos de crecimiento poblacional

Tomando en cuenta los datos faltantes, hacer filtros para seleccionar una muestra mas pequeña

# 2023 aun no tiene registro en ODA
datos_paper %>% filter(is.na(DT.ODA.ALLD.CD), !year %in% c(2023)) ## SS (South Sudan) y ZW (Zimbabwe) faltantes de ODA y GOB indicators
datos_paper %>% filter(!iso2c %in% c('SS', 'ZW')) %>% filter(is.na(CC.EST)) %>% group_by(year) %>% summarise(times = n())
# para años 1995, 1997, 1999, 2001 y 2023 no hay registros de GOB
# 1996, 1998, 2000, 2002 and 2003 tiene algunos paises sin datos
datos_paper %>% arrange(year) %>% filter(!iso2c %in% c('SS', 'ZW'), !year %in% c(1995, 1997, 1999, 2001, 2023)) %>%
                filter(is.na(CC.EST)) # FM (Micronesia), KI (Kiribati) y TL (Timor-Leste) no tiene GOB in en estos años 
                                      # tambien CV (Cabo Verde) and SB (Solomon Islands) no registro GOB en 2000 - 2003
datos_paper %>% arrange(iso2c) %>% 
          filter(!iso2c %in% c('SS','ZW','BT','ER','GW','KP','LB','NG','PS','SO','VU','FM','KI','TL','CB','CV','SB'), 
                 !year %in% c(1995, 1996, 1997, 1998, 1999, 2000, 2001, 2023)) %>%
          select(iso2c, year, hdi,
                 all_of(gsub("_", ".", gob_indicators))
                 )
# Ver datos aplicando los filtros determinados en las busquedas pasadas
# antes del 2001 suele tener informacion faltante
# BT (Bhutan), ER (Eritrea), GW (Guinea-Bissau), KP (North Korea), LB (Lebanon), NG (Nigeria), PS (Palestine), SO (Somalia), VU (Vanuatu) son paises # sin registro de hdi
datos_paper %>% 
        filter(!iso2c %in% c('SS', 'ZW', 'BT', 'ER', 'GW', 'KP', 'LB', 'NG', 'PS', 'SO', 'VU', 'FM', 'KI', 'TL', 'CV', 'SB','SY'),
               !year %in% c(1995, 1996, 1997, 1998, 1999, 2000, 2023)) %>%
        select(iso2c, year, hdi, DT.ODA.ALLD.CD, DT.ODA.ALLD.KD, DT.ODA.ODAT.CD, DT.ODA.ODAT.KD, DT.ODA.ODAT.PC.ZS, 
               NY.GDP.PCAP.CN, NY.GDP.PCAP.CD, SP.POP.GROW, all_of(gsub("_", ".", gob_indicators)), 
               ) %>%
        filter(is.na(CC.EST))
datos_paper %>% 
        filter(!iso2c %in% c('SS', 'ZW', 'BT', 'ER', 'GW', 'KP', 'LB', 'NG', 'PS', 'SO', 'VU', 'FM', 'KI', 'TL', 'CV', 'SB','SY'),
               !year %in% c(1995, 1996, 1997, 1998, 1999, 2000, 2023)) %>%
        select(iso2c, year, hdi, DT.ODA.ALLD.CD, DT.ODA.ALLD.KD, DT.ODA.ODAT.CD, DT.ODA.ODAT.KD, DT.ODA.ODAT.PC.ZS, 
               NY.GDP.PCAP.CN, NY.GDP.PCAP.CD, SP.POP.GROW, all_of(gsub("_", ".", gob_indicators)), 
               )
vis_dat(datos_paper %>% 
        filter(!iso2c %in% c('SS', 'ZW', 'BT', 'ER', 'GW', 'KP', 'LB', 'NG', 'PS', 'SO', 'VU', 'FM', 'KI', 'TL', 'CV', 'SB','SY'),
               !year %in% c(1995, 1996, 1997, 1998, 1999, 2000, 2023)) %>%
        select(iso2c, year, hdi, DT.ODA.ALLD.CD, DT.ODA.ALLD.KD, DT.ODA.ODAT.CD, DT.ODA.ODAT.KD, DT.ODA.ODAT.PC.ZS, 
               NY.GDP.PCAP.CN, NY.GDP.PCAP.CD, SP.POP.GROW, all_of(gsub("_", ".", gob_indicators)), 
               ))

vis_dat(datos_paper %>% 
        filter(!iso2c %in% c('SS', 'ZW', 'BT', 'ER', 'GW', 'KP', 'LB', 'NG', 'PS', 'SO', 'VU', 'FM', 'KI', 'TL', 'CV', 'SB','SY'),
               !year %in% c(1995, 1996, 1997, 1998, 1999, 2000, 2001, 2023)) %>%
        select(iso2c, year, hdi, DT.ODA.ALLD.CD, DT.ODA.ALLD.KD, DT.ODA.ODAT.CD, DT.ODA.ODAT.KD, DT.ODA.ODAT.PC.ZS, 
               NY.GDP.PCAP.CN, NY.GDP.PCAP.CD, SP.POP.GROW, all_of(gsub("_", ".", gob_indicators)), 
               ))

De 2232 observaciones reducimos a 1098 (2002 hasta 2019) o a 1260 (2002 hasta 2022)

Aplicar Operador diferencia

datos_model <- datos_paper %>% 
        filter(!iso2c %in% c('SS', 'ZW', 'BT', 'ER', 'GW', 'KP', 'LB', 'NG', 'PS', 'SO', 'VU', 'FM', 'KI', 'TL', 'CV', 'SB', 'SY'),
               !year %in% c(1995, 1996, 1997, 1998, 1999, 2000, 2023)) %>%
        select(iso2c, year, hdi, DT.ODA.ALLD.CD, DT.ODA.ALLD.KD, DT.ODA.ODAT.CD, DT.ODA.ODAT.KD, DT.ODA.ODAT.PC.ZS, 
               NY.GDP.PCAP.CN, NY.GDP.PCAP.CD, SP.POP.GROW, all_of(gsub("_", ".", gob_indicators))
               )

datos_model <- datos_model %>% arrange(iso2c, year) %>% 
        mutate(hdi_diff = hdi - lag(hdi), 
               NY.GDP.PCAP.CN_diff = NY.GDP.PCAP.CN - lag(NY.GDP.PCAP.CN), 
               NY.GDP.PCAP.CD_diff = NY.GDP.PCAP.CD - lag(NY.GDP.PCAP.CD),
               DT.ODA.ALLD.CD_diff = DT.ODA.ALLD.CD - lag(DT.ODA.ALLD.CD),
               DT.ODA.ODAT.PC.ZS_diff = DT.ODA.ODAT.PC.ZS - lag(DT.ODA.ODAT.PC.ZS)) %>%
        filter(!year %in% c(2001))

vis_dat(datos_model)

Modelos Regresion Lineal

Probando modelos sencillos, regresion lineal, Minimos cuadrados, datos panel, HDI o GDP o sus differecias

HDI <- ODA.ALL

HDI = ODA.ALL + CC + GE + PV + RQ + RL + VA + POP.GROW

summary(lm(hdi ~ DT.ODA.ALLD.CD + CC.EST + GE.EST + PV.EST + RQ.EST + RL.EST + VA.EST + SP.POP.GROW, data=datos_model))
## 
## Call:
## lm(formula = hdi ~ DT.ODA.ALLD.CD + CC.EST + GE.EST + PV.EST + 
##     RQ.EST + RL.EST + VA.EST + SP.POP.GROW, data = datos_model)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.24376 -0.05200 -0.00272  0.05310  0.48585 
## 
## Coefficients:
##                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)     6.624e-01  6.715e-03  98.642  < 2e-16 ***
## DT.ODA.ALLD.CD  6.180e-12  2.369e-12   2.609 0.009195 ** 
## CC.EST         -7.398e-02  9.048e-03  -8.176 7.13e-16 ***
## GE.EST          1.115e-01  9.458e-03  11.790  < 2e-16 ***
## PV.EST          1.501e-02  3.885e-03   3.864 0.000117 ***
## RQ.EST         -6.752e-04  9.569e-03  -0.071 0.943761    
## RL.EST          2.889e-02  1.046e-02   2.762 0.005834 ** 
## VA.EST         -1.363e-02  4.712e-03  -2.892 0.003895 ** 
## SP.POP.GROW    -3.803e-02  2.400e-03 -15.849  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.08028 on 1251 degrees of freedom
## Multiple R-squared:  0.427,  Adjusted R-squared:  0.4234 
## F-statistic: 116.6 on 8 and 1251 DF,  p-value: < 2.2e-16
# Todas las variables son significativas al 99% excepto Regulatory Quality

HDI <- ODA.PC

HDI = ODA.PC + CC + GE + PV + RQ + RL + VA + POP.GROW

summary(lm(hdi ~ DT.ODA.ODAT.PC.ZS + CC.EST + GE.EST + PV.EST + RQ.EST + RL.EST + VA.EST + SP.POP.GROW, data=datos_model))
## 
## Call:
## lm(formula = hdi ~ DT.ODA.ODAT.PC.ZS + CC.EST + GE.EST + PV.EST + 
##     RQ.EST + RL.EST + VA.EST + SP.POP.GROW, data = datos_model)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.23281 -0.05385 -0.00335  0.05343  0.43779 
## 
## Coefficients:
##                     Estimate Std. Error t value Pr(>|t|)    
## (Intercept)        6.449e-01  7.101e-03  90.823  < 2e-16 ***
## DT.ODA.ODAT.PC.ZS  2.411e-04  3.386e-05   7.120 1.81e-12 ***
## CC.EST            -9.071e-02  9.130e-03  -9.935  < 2e-16 ***
## GE.EST             1.205e-01  9.294e-03  12.968  < 2e-16 ***
## PV.EST             6.662e-03  3.532e-03   1.886 0.059520 .  
## RQ.EST             8.239e-03  9.499e-03   0.867 0.385940    
## RL.EST             2.907e-02  1.017e-02   2.857 0.004342 ** 
## VA.EST            -1.744e-02  4.658e-03  -3.745 0.000189 ***
## SP.POP.GROW       -3.805e-02  2.344e-03 -16.231  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.07891 on 1251 degrees of freedom
## Multiple R-squared:  0.4464, Adjusted R-squared:  0.4428 
## F-statistic: 126.1 on 8 and 1251 DF,  p-value: < 2.2e-16
# Todas las variables son significativas al 99% excepto Regulatory Quality

GPD.PC <- ODA.ALL

GPD.PC = ODA.ALL + CC + GE + PV + RQ + RL + VA + POP.GROW

summary(lm(NY.GDP.PCAP.CD ~ DT.ODA.ALLD.CD + CC.EST + GE.EST + PV.EST + RQ.EST + RL.EST + VA.EST + SP.POP.GROW, data=datos_model))
## 
## Call:
## lm(formula = NY.GDP.PCAP.CD ~ DT.ODA.ALLD.CD + CC.EST + GE.EST + 
##     PV.EST + RQ.EST + RL.EST + VA.EST + SP.POP.GROW, data = datos_model)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2084.8  -611.6  -147.7   441.7  4237.9 
## 
## Coefficients:
##                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)     2.461e+03  7.400e+01  33.256  < 2e-16 ***
## DT.ODA.ALLD.CD  4.006e-09  2.611e-08   0.153 0.878095    
## CC.EST         -9.011e+01  9.972e+01  -0.904 0.366385    
## GE.EST          7.899e+02  1.042e+02   7.579 6.77e-14 ***
## PV.EST          1.556e+02  4.281e+01   3.634 0.000291 ***
## RQ.EST          9.097e+01  1.055e+02   0.863 0.388498    
## RL.EST          1.399e+02  1.153e+02   1.214 0.225078    
## VA.EST         -2.908e+02  5.193e+01  -5.600 2.63e-08 ***
## SP.POP.GROW    -2.049e+02  2.645e+01  -7.747 1.93e-14 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 884.7 on 1251 degrees of freedom
## Multiple R-squared:  0.286,  Adjusted R-squared:  0.2814 
## F-statistic: 62.63 on 8 and 1251 DF,  p-value: < 2.2e-16
# Todas las variables son significativas al 99% excepto ODA.ALL, Control of Corruption, Regulatory Quality y Rule of Law

GPD.PC <- ODA.PC

GPD.PC = ODA.PC + CC + GE + PV + RQ + RL + VA + POP.GROW

summary(lm(NY.GDP.PCAP.CD ~ DT.ODA.ODAT.PC.ZS + CC.EST + GE.EST + PV.EST + RQ.EST + RL.EST + VA.EST + SP.POP.GROW, data=datos_model))
## 
## Call:
## lm(formula = NY.GDP.PCAP.CD ~ DT.ODA.ODAT.PC.ZS + CC.EST + GE.EST + 
##     PV.EST + RQ.EST + RL.EST + VA.EST + SP.POP.GROW, data = datos_model)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -1876.5  -586.5  -166.4   364.0  4382.8 
## 
## Coefficients:
##                    Estimate Std. Error t value Pr(>|t|)    
## (Intercept)       2152.1327    76.8078  28.020  < 2e-16 ***
## DT.ODA.ODAT.PC.ZS    3.5326     0.3663   9.644  < 2e-16 ***
## CC.EST            -314.1459    98.7580  -3.181   0.0015 ** 
## GE.EST             887.1625   100.5340   8.825  < 2e-16 ***
## PV.EST              91.4305    38.2095   2.393   0.0169 *  
## RQ.EST             252.0015   102.7496   2.453   0.0143 *  
## RL.EST              83.6278   110.0482   0.760   0.4474    
## VA.EST            -343.0970    50.3834  -6.810 1.51e-11 ***
## SP.POP.GROW       -215.6641    25.3596  -8.504  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 853.6 on 1251 degrees of freedom
## Multiple R-squared:  0.3354, Adjusted R-squared:  0.3311 
## F-statistic: 78.91 on 8 and 1251 DF,  p-value: < 2.2e-16
# Todas las variables son significativas al 99% excepto Rule of Law

Se revisara las relaciones entre las variables graficamente

Visualizacion de datos

HDI

my_plot  <- list()

for (col in c('DT.ODA.ALLD.CD', 'DT.ODA.ODAT.PC.ZS', 'DT.ODA.ALLD.CD_diff', 'DT.ODA.ODAT.PC.ZS_diff', 
              'CC.EST', 'GE.EST', 'PV.EST', 'RQ.EST', 'RL.EST', 'VA.EST')) {
  my_plot[[col]] <- plot_ly(x = datos_model[[col]], y = datos_model[[3]], type = 'scatter', mode = 'markers', name = col)  
}
subplot(my_plot[1:4], nrows = 2, margin = 0.05) %>% layout(title = 'HDI vs ODA')
subplot(my_plot[5:10], nrows = 2, margin = 0.05) %>% layout(title = 'HDI vs GOB')

HDI diff

my_plot  <- list()

for (col in c('DT.ODA.ALLD.CD', 'DT.ODA.ODAT.PC.ZS', 'DT.ODA.ALLD.CD_diff', 'DT.ODA.ODAT.PC.ZS_diff', 
              'CC.EST', 'GE.EST', 'PV.EST', 'RQ.EST', 'RL.EST', 'VA.EST')) {
  my_plot[[col]] <- plot_ly(x = datos_model[[col]], y = datos_model[[48]], type = 'scatter', mode = 'markers', name = col)  
}
subplot(my_plot[1:4], nrows = 2, margin = 0.05) %>% layout(title = 'HDI diff vs ODA')
subplot(my_plot[5:10], nrows = 2, margin = 0.05) %>% layout(title = 'HDI diff vs GOB')

GPD.PC

my_plot  <- list()
for (col in c('DT.ODA.ALLD.CD', 'DT.ODA.ODAT.PC.ZS', 'DT.ODA.ALLD.CD_diff', 'DT.ODA.ODAT.PC.ZS_diff', 
              'CC.EST', 'GE.EST', 'PV.EST', 'RQ.EST', 'RL.EST', 'VA.EST')) {
  my_plot[[col]] <- plot_ly(x = datos_model[[col]], y = datos_model[[10]], type = 'scatter', mode = 'markers', name = col)  
}
subplot(my_plot[1:4], nrows = 2, margin = 0.05) %>% layout(title = 'GDP.PC vs ODA')
subplot(my_plot[5:10], nrows = 2, margin = 0.05) %>% layout(title = 'GDP.PC vs GOB')

GPD.PC diff

my_plot  <- list()
for (col in c('DT.ODA.ALLD.CD', 'DT.ODA.ODAT.PC.ZS', 'DT.ODA.ALLD.CD_diff', 'DT.ODA.ODAT.PC.ZS_diff', 
              'CC.EST', 'GE.EST', 'PV.EST', 'RQ.EST', 'RL.EST', 'VA.EST')) {
  my_plot[[col]] <- plot_ly(x = datos_model[[col]], y = datos_model[[50]], type = 'scatter', mode = 'markers', name = col)  
}
subplot(my_plot[1:4], nrows = 2, margin = 0.05) %>% layout(title = 'GDP.PC diff vs ODA')
subplot(my_plot[5:10], nrows = 2, margin = 0.05) %>% layout(title = 'GDP.PC diff vs GOB')

No se ve una relacion clara, hay tanto paises con punteos altos y bajos de GOB que tienen tanto HID altos o bajos Quiza puede verse una leve relacion de mayor punteo en GOB acompañado de mejor punteo den HDI Los datos de GPD si muestran una relacion positiva con el HDI visto en las graficas

Se realizara el mismo proceso con el crecimiento o decrecimiento de HDI anual (no se perderan datos al calcular la diferencia porque se añade el año 2001 en la seleccion)

Visualizacion de Datos historia

Viendo la historia de las variables en el tiempo (por pais)

datos_model %>% filter(iso2c == 'AF') %>% plot_ly(x = ~year) %>% 
  add_trace(y = ~hdi, type = 'scatter', mode = 'lines+markers', name = 'hdi') %>% 
  add_trace(y = ~NY.GDP.PCAP.CD / 1000, type = 'scatter', mode = 'lines+markers', name = 'gdp.pc') %>% 
  add_trace(y = ~DT.ODA.ALLD.CD / 10000000000, type = 'scatter', mode = 'lines+markers', name = 'ODA.ALL')  %>% 
  add_trace(y = ~DT.ODA.ODAT.PC.ZS / 1000, type = 'scatter', mode = 'lines+markers', name = 'ODA.PC')  %>% 
  add_trace(y = ~CC.EST, type = 'scatter', mode = 'lines+markers', name = 'CC') %>% 
  add_trace(y = ~GE.EST, type = 'scatter', mode = 'lines+markers', name = 'GE')  %>% 
  add_trace(y = ~PV.EST, type = 'scatter', mode = 'lines+markers', name = 'PV')  %>% 
  add_trace(y = ~RQ.EST, type = 'scatter', mode = 'lines+markers', name = 'RQ')  %>% 
  add_trace(y = ~RL.EST, type = 'scatter', mode = 'lines+markers', name = 'RL')  %>% 
  add_trace(y = ~VA.EST, type = 'scatter', mode = 'lines+markers', name = 'VA')

Modelos Efectos fijos

HDI <- ODA.ALL

summary(lm(hdi ~ DT.ODA.ALLD.CD + CC.EST + GE.EST + PV.EST + RQ.EST + RL.EST + VA.EST + SP.POP.GROW + iso2c, data=datos_model))
## 
## Call:
## lm(formula = hdi ~ DT.ODA.ALLD.CD + CC.EST + GE.EST + PV.EST + 
##     RQ.EST + RL.EST + VA.EST + SP.POP.GROW + iso2c, data = datos_model)
## 
## Residuals:
##       Min        1Q    Median        3Q       Max 
## -0.142667 -0.023725  0.003785  0.024445  0.095473 
## 
## Coefficients:
##                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)     4.622e-01  1.632e-02  28.319  < 2e-16 ***
## DT.ODA.ALLD.CD  2.089e-11  1.533e-12  13.623  < 2e-16 ***
## CC.EST         -8.331e-03  6.874e-03  -1.212 0.225736    
## GE.EST          7.539e-03  6.801e-03   1.108 0.267883    
## PV.EST         -4.462e-03  2.915e-03  -1.531 0.126074    
## RQ.EST          2.094e-02  6.706e-03   3.122 0.001838 ** 
## RL.EST          3.256e-02  7.769e-03   4.191 2.99e-05 ***
## VA.EST          6.926e-03  5.106e-03   1.356 0.175199    
## SP.POP.GROW    -5.590e-03  1.881e-03  -2.972 0.003014 ** 
## iso2cAO         1.452e-01  1.323e-02  10.971  < 2e-16 ***
## iso2cBD         1.144e-01  1.316e-02   8.694  < 2e-16 ***
## iso2cBF        -5.751e-02  1.452e-02  -3.962 7.88e-05 ***
## iso2cBI         2.206e-03  1.263e-02   0.175 0.861349    
## iso2cBJ         5.345e-02  1.480e-02   3.611 0.000318 ***
## iso2cBO         2.491e-01  1.447e-02  17.214  < 2e-16 ***
## iso2cCD         1.970e-02  1.126e-02   1.749 0.080562 .  
## iso2cCF        -2.202e-02  1.323e-02  -1.664 0.096280 .  
## iso2cCG         1.933e-01  1.344e-02  14.391  < 2e-16 ***
## iso2cCI         6.188e-02  1.302e-02   4.753 2.25e-06 ***
## iso2cCM         1.141e-01  1.282e-02   8.897  < 2e-16 ***
## iso2cDJ         3.787e-02  1.506e-02   2.514 0.012073 *  
## iso2cEG         2.057e-01  1.464e-02  14.055  < 2e-16 ***
## iso2cET        -5.384e-02  1.320e-02  -4.078 4.85e-05 ***
## iso2cGH         9.150e-02  1.588e-02   5.763 1.05e-08 ***
## iso2cGM         3.844e-02  1.516e-02   2.535 0.011357 *  
## iso2cGN         3.450e-02  1.299e-02   2.657 0.008001 ** 
## iso2cHN         1.750e-01  1.409e-02  12.418  < 2e-16 ***
## iso2cHT         1.254e-01  1.387e-02   9.037  < 2e-16 ***
## iso2cIN         8.076e-02  1.609e-02   5.019 5.99e-07 ***
## iso2cJO         2.433e-01  1.685e-02  14.440  < 2e-16 ***
## iso2cKE         8.970e-02  1.338e-02   6.702 3.17e-11 ***
## iso2cKG         2.505e-01  1.410e-02  17.770  < 2e-16 ***
## iso2cKH         1.271e-01  1.428e-02   8.900  < 2e-16 ***
## iso2cKM         1.514e-01  1.523e-02   9.942  < 2e-16 ***
## iso2cLA         1.675e-01  1.533e-02  10.928  < 2e-16 ***
## iso2cLK         2.759e-01  1.683e-02  16.389  < 2e-16 ***
## iso2cLR         6.486e-02  1.372e-02   4.727 2.55e-06 ***
## iso2cLS         4.852e-02  1.707e-02   2.842 0.004566 ** 
## iso2cMA         1.553e-01  1.552e-02  10.004  < 2e-16 ***
## iso2cMG         5.928e-02  1.414e-02   4.191 2.98e-05 ***
## iso2cML        -4.627e-02  1.380e-02  -3.353 0.000825 ***
## iso2cMM         1.313e-01  1.426e-02   9.206  < 2e-16 ***
## iso2cMR         1.027e-01  1.405e-02   7.310 4.88e-13 ***
## iso2cMW         2.355e-02  1.489e-02   1.582 0.113913    
## iso2cMZ        -3.347e-02  1.331e-02  -2.514 0.012084 *  
## iso2cNE        -8.701e-02  1.347e-02  -6.459 1.53e-10 ***
## iso2cNI         1.961e-01  1.447e-02  13.554  < 2e-16 ***
## iso2cNP         1.054e-01  1.449e-02   7.277 6.17e-13 ***
## iso2cPG         9.099e-02  1.393e-02   6.532 9.58e-11 ***
## iso2cPH         2.229e-01  1.554e-02  14.346  < 2e-16 ***
## iso2cPK         4.254e-02  1.301e-02   3.270 0.001106 ** 
## iso2cRW         3.929e-02  1.670e-02   2.353 0.018786 *  
## iso2cSD         8.421e-02  1.253e-02   6.722 2.77e-11 ***
## iso2cSL         1.377e-02  1.417e-02   0.972 0.331360    
## iso2cSN         1.830e-02  1.522e-02   1.202 0.229580    
## iso2cST         1.575e-01  1.640e-02   9.606  < 2e-16 ***
## iso2cSZ         1.063e-01  1.705e-02   6.234 6.31e-10 ***
## iso2cTD        -1.773e-02  1.271e-02  -1.395 0.163297    
## iso2cTG         8.637e-02  1.437e-02   6.011 2.45e-09 ***
## iso2cTJ         2.438e-01  1.354e-02  18.005  < 2e-16 ***
## iso2cTN         2.456e-01  1.679e-02  14.626  < 2e-16 ***
## iso2cTZ         2.165e-02  1.356e-02   1.597 0.110578    
## iso2cUG         3.464e-02  1.395e-02   2.483 0.013151 *  
## iso2cUZ         2.925e-01  1.448e-02  20.197  < 2e-16 ***
## iso2cVN         2.103e-01  1.596e-02  13.181  < 2e-16 ***
## iso2cWS         2.167e-01  2.069e-02  10.472  < 2e-16 ***
## iso2cYE         4.979e-02  1.205e-02   4.134 3.82e-05 ***
## iso2cZM         9.126e-02  1.459e-02   6.255 5.53e-10 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.03469 on 1192 degrees of freedom
## Multiple R-squared:  0.898,  Adjusted R-squared:  0.8923 
## F-statistic: 156.7 on 67 and 1192 DF,  p-value: < 2.2e-16
#plm(hdi ~ DT.ODA.ALLD.CD + CC.EST + GE.EST + PV.EST + RQ.EST + RL.EST + VA.EST + SP.POP.GROW, data=datos_model,
#           index = c("iso2c", "year"), model = "within")